4 – Project execution on the NCS

This chapter summarises experience acquired from completed and ongoing projects on the NCS, with particular attention paid to the post-2013 period.

Cost data for the projects have been taken from Proposition (Bill) 1 S to the Storting, which is drawn up annually by the MPE as part of the input to the national budget. This presents the project’s PDO estimate, the latest updated cost estimate, and cost trends since the PDO and over the previous year. The figure presented is the project’s total costs, which include possible gain/loss on changes to the exchange rates assumed in the PDO.

Costs estimates in this report have been inflation-adjusted to 2019 value in line with the consumer price index. Operators report to the MPE up to the start of production. Work may still be outstanding in a project after the field has come on stream – related to drilling wells, for example. The figures in the national budget are therefore the latest cost estimate, which could vary from the final development cost.

Planning data are taken from the PDO and the NPD’s fact pages. The actual start date is compared with the planned time stated in the PDO. The latter is often specified as a month rather than a day. In those cases where no specific date is stated, the base data is attributed to the last day of the month.

Developments in field reserves are taken from the NPD’s resource accounts, based on reporting to the revised national budget for 2019 (reported in the autumn of 2018).

4.1 Overview of sanctioned development projects

PDOs for 36 projects were approved in 2007-13. The corresponding figure for 2013-18 was 30. Figure 3 presents an overview of the types of development concepts chosen for these projects. Most were subsea developments, followed by fixed and floating facilities.

The distribution of development concepts is comparable for the two periods. Total estimated costs came to just over NOK 470 billion for both, and break down more or less equally between the various development concepts. See figures 4 and 5. Fixed facilities account for roughly 45 per cent of the estimates, with subsea developments and floating facilities representing about a quarter each. *

Figure 3 Projects by development concept. A total of 66 projects had an approved PDO/PIO in 2007-18.

* Most projects incorporate subsea installations and wells in their development concept. In this report, the fixed and floating facility category covers fields developed with such installations (but which may also feature subsea wells/equipment). The subsea development category covers fields where the seabed facilities are tied back to existing infrastructure, and the well category encompasses a project limited to wells. The costs specified in this report are the total figure for a project, and are not broken down by discipline.

Big variations may exist within each category in the functionality, complexity and cost of development concepts. Fixed facilities, for example, include phase 1 of Johan Sverdrup, with four such facilities and a PDO estimate of almost NOK 130 billion, as well as Oseberg west flank 2, an unmanned wellhead platform estimated at about NOK 8.5 billion in the PDO. The floaters cover various hull types – Aasta Hansteen is a Spar, Goliat has a circular floating production, storage and offloading (FPSO) unit, Knarr and Johan Castberg have ship-shaped FPSOs and Gjøa is a semi-submersible.

Most of the projects are newbuilds, but some are also large modifications. Njord Future involves upgrading the existing Njord A and B facilities. Yme New Development is based on reusing equipment left from the earlier project on this field, along with readying the Maersk Inspirer drilling rig for production.

The subsea development category also embraces big variations in concepts and complexity, with the number and type of wells and pipelines as well as the distance from and scope of work on the host facility differing between the projects. Investment estimates vary from about NOK 1.5 billion for Skogul and Hyme to just under NOK 20 billion for Snorre Subsea Expansion (SEP).

Fifty-three of the projects had come on stream in 2019, with 12 still under development. One – Yme, with its PDO approved in 2007 – was terminated without completing the project. A new Yme development was approved in 2018. This is still in progress and forms part of the base data.

Figure 4 Projects by development concept and period. Thirty-six projects were sanctioned in 2007-12 and 30 in 2013-18.

Figure 5 Distribution of PDO cost estimates by development concept and period. Planned investment for projects in 2007-12 totalled NOK 474 billion in 2019 value. The overall cost estimate for 2013-18 was NOK 476 billion in 2019 value.

An overview of projects covered in this report and their operator is provided at the end of the report. Equinor accounted for about half of the developments and of all planned investments. The remaining half breaks down between 16 operators who were responsible for one-three projects each.

4.2 Cost developments

All costs underpinning the decision taken at the PDO/PAD stage are estimates. These will take account of uncertainties in the project, and therefore lie within an interval expressing a certain degree of confidence. More detailed engineering is required to firm up the estimates. How firm the latter must be before a project is sanctioned will always be a matter of judgement.

Licensees on the NCS normally require that estimated costs have a maximum uncertainty of +/- 20 per cent within an 80 per cent confidence interval at the PDO. This means that, if a given project is repeated many times, estimated costs would be within the +/- 20 per cent uncertainty range in eight out of 10 cases. A project where costs increase or decrease by less than 20 per cent of the PDO estimate is thereby considered to have been implemented to budget.

At the PDO stage, the licensees prepare a master control estimate (MCE). Their project organisation monitors cost developments (and the plan) throughout the execution phase. Updated cost estimates and plans – known as the current control estimate (CCE) – are prepared regularly for the projects, and the latest of these is used as the basis for reporting to Proposition 1 S. Monthly project reports sum up progress and cost developments compared with the latest CCE update and the MCE.

Projects which came on stream in the autumn of 2019 (Johan Sverdrup phase 1, Utgard, Valhall flank west) are included in this report, but will secure a new and more updated cost estimate in the 2021 national budget. The base data also include a number of projects not yet on stream. Their costs estimates are thereby more uncertain and could change.

Most projects are implemented without cost overruns

Figures for the 66 projects with approved PDOs in 2007-18 show that 83 per cent were completed either within the uncertainty range in the PDO estimate or below. See figure 6. In the order of 73 per cent of the projects were completed in line with the PDO estimate. Just under 17 per cent had cost overruns and 11 per cent saw costs reduced by more than 20 per cent. That includes projects which are still ongoing.

Viewed overall, the projects saw their costs rise by about eight per cent (roughly NOK 75 billion) from the PDO estimate. The category of fixed and floating facilities made the largest contribution to an overall increase in costs for the projects. See figure 7.

Figure 6 shows the number of projects completed within the PDO uncertainty range (+/- 20 per cent) and the number completed with costs increases/decreases greater than 20 per cent. Total of 66 projects. The figure also includes projects which are not yet completed.

Figure 7 Cost developments for 66 projects in NOK billion in 2019 value by development concept. Each dot represents a project, and its colour indicates the development concept. The squares represent the total cost trend for projects in a development category. Johan Sverdrup phase 1 is not presented as an individual project because of the size of the investment, but has been incorporated when calculating the overall cost trend for the fixed facility category.

Project execution improved from the first six-year period to the second

A comparison of projects approved in 2007-12 and 2013-18 respectively shows that execution in more recent years was better than before. This is illustrated by figures 8 and 9. Projects approved in recent years have hit their cost estimate better, and fewer have had overruns. The final status of developments yet to be completed remains uncertain.

Overall, costs for pre-2013 projects increased by about NOK 115 billion or 24 per cent from the PDO estimate. Eight of the projects ended up with cost overruns – Vega and Vega South, Valhall Redevelopment, Skarv, Yme, Goliat, Brynhild, Jette and Martin Linge. Of these, Martin Linge is still not on stream and Yme was terminated without being completed. Two projects in this period – Troll P-12 and Troll B gas injection – witnessed a cost reduction of more than 20 per cent.

The post-2013 projects saw an overall cost reduction of NOK 40 billion or eight per cent from the PDO estimate. Three had cost overruns – Flyndre, Varg gas export and Njord Future. No less than six competed projects – Oseberg west flank 2, Maria, Ekofisk 2/4, Sverdrup construction stage 1, Edvard Grieg oil pipeline and Rutil in Gullfaks Rimfaks Valley – reduced costs by more than 20 per cent.

Figure 8 Cost trends for development projects (percentage change from the PDO estimate) compared with the year of official PDO/PIO approval.

Figure 9 Overall cost trend for development projects by development concept. The grey area shows the normal +/- 20 per cent uncertainty range in the cost estimates.

4.2.1 Fixed and floating facilities

This section looks in more detail at cost trends for developments based on fixed or floating facilities. A total of 21 projects fall into that category. Figure 10 provides an overview of their cost performance in per cent.

For the period as a whole, 71 per cent of the projects (15 of 21) ended up within or below the uncertainty range for costs. Six had cost overruns, and one reduced costs by more than 20 per cent.

Most projects with overruns had their PDO approved before 2013. Since 2013, only one project has had overruns – Nord Future, currently under development. Johan Sverdrup phase 1 had savings of 24 per cent (about NOK 31 billion). Several of the projects with large overruns were covered in the NPD’s 2013 report.

Topside costs for many projects are more than 20 per cent above the estimate, but the total cost may nevertheless remain within the uncertainty range. This will often be the case if the project manages to stay on schedule and costs for other elements are below the estimate. Examples are Gina Krog, Ivar Aasen and Edvard Grieg.

Figure 10 Cost changes in per cent for development projects involving fixed or floating facilities.

Major modifications and upgrades of existing facilities could help to increase complexity, since  facility and the Njord B storage ship. Converted tankers were chosen as storage ships for both Gina Krog and Martin Linge rather than building new vessels. Both were significantly delayed.

Estimates for the steel jackets on fixed platforms are normally good. Floater hulls have experienced cost increases in some projects. That applies to both Goliat and Aasta Hansteen, where the hulls were larger and more complex than with similar the condition of the facility may be uncertain. Njord Future involves upgrades and modifications to both the Njord A productionearlier concepts. The hulls for Skarv (apart from the turret) and Gjøa were cheaper than estimated.

Erroneous assumptions about the krone exchange rate in their PDO contributed to a substantial cost increase for several projects. Aasta Hansteen, Martin Linge and Johan Sverdrup phase 1 suffered foreign exchange losses of several billion kroner. Despite negative currency effects for several of the projects approved in 2013 and later, these had an overall cost reduction of about NOK 40 billion.

4.2.2 Subsea developments

This section takes a closer look at cost trends for subsea developments. A total of 38 projects fall into that category. Figure 11 presents an overview of their cost performance in per cent.

Ninety per cent of the projects ended up with costs within the PDO uncertainty range or below. Seventy-nine per cent (30 of 38) were completed to the budget presented in the PDO. Four (10.5 per cent) had cost overruns and four were below the uncertainty range. The proportion completed to the PDO estimate was larger than for fixed and floating facilities.

A further improvement has also been achieved in recent years. Most projects with a PDO approved post-2013 are set to end up with costs lower than estimated. Only one of these 16 – Flyndre – has so far had a cost increase above the PDO range, and no less than three – Maria, Rutil in Gullfaks Rimfaks Valley and Ekofisk 2/4 VC – have seen costs reduced below it.

Figure 11 Cost changes in per cent for development projects involving subsea installations.

In connection with the cost trends reported annually in Proposition 1 S, the MPE provides a brief summary of the reasons for any changes to cost estimates. A review of this summary for subsea developments shows no clear reasons for the cost variations.

Work on a producing facility calls for good planning to limit the impact on day-to-day operations. Estimating the scope of the modifications can be demanding. Nevertheless, the NPD has not found this to be an area where the projects fail to succeed more often than others. However, three of those with cost overruns – Jette, Flyndre and Brynhild – reported modifications as a source of overruns.

A common denominator for the four projects which reduced costs by more than 20 per cent is that drilling proved cheaper than the PDO estimate. Maria, Rutil in Gullfaks Rimfaks Valley and Ekofisk 2/4 VC all benefited from industry improvements related to drilling efficiency in recent years.

4.3 Schedule

Figure 12 shows the deviation between scheduled and actual start-up for completed projects. A majority came on stream within a reasonable time, but the overall average delay in relation to the schedule was about 3.5 months.

Some individual projects in particular took significantly longer. Martin Linge is not included in the base data since it remains uncompleted. Nor has Yme been included since it was halted in December 2012 without being finished. That corresponded at the time to a delay of about four years from the planned start-up date.

The average delay for fixed and floating facilities in 2007-18 was just under seven months. Four of the projects – Ekofisk South, Eldfisk II, Johan Sverdrup phase 1 and Valhall flank west – were completed before or on schedule. Projects approved in 2013 and later did better than in the preceding period, with an average delay of 3.5 months.

Figure 12 Number of days of delay by project and development concept. Ongoing projects are not included. Projects with a negative number of days have come on stream ahead of schedule.

Subsea developments experienced an average delay of just under two months. Projects approved in 2013 and later again did better than in the preceding period, being completed on average two months ahead of schedule rather than three months behind.

When developing a field, many interdependent activities must be planned and executed. Great uncertainty could exist over the duration of these and over delays which might propagate further through the project.

A planning risk analysis is normally carried out to identify how different risks could affect the schedule. That forms the basis for establishing a probability distribution for when the project could be completed and a best estimate for coming on stream.

Figure 13 Variation from the planned start-up date for completed projects compared with changes in costs for projects with a PDO approved in 2007-18.

In order to meet the expected start-up date when building facilities with a large amount of offshore work (such as topsides installation and hook-up), it is generally important to ensure that departure from the yard occurs in spring or early summer. Much of the installation work is weather-sensitive and must occur between April and September. A few months of delay from the yard could thereby put the project a whole year behind schedule.

Subsea developments also depend on a sufficiently good weather window for carrying out certain activities, but are less vulnerable if some deliveries are delayed.

Figure 13 shows that a relationship exists between delays and cost developments. The spread in the numbers is relatively large. This could be because cost increases and delays occur only in parts of the work, while possible savings are made elsewhere. Contractual relationships as well as the operator’s project portfolio and opportunities to swap installation activities and vessels are also significant here.

4.4 Regularity

Development projects may often be assessed on the basis of development costs and execution time. However, it is also important that the facilities can be operated in a good way and that production takes place as expected.

Regularity, field lifetime and the level of safety can be affected if equipment which has been fabricated and installed fails to meet the desired quality. Based on a review of reported production from the fields as well as information about operating experience provided in the annual status reports, thNPD’s assessment is that projects with cost overruns and delays have an increased risk of lower uptime after coming on stream. This could be because production facilities have not been fully completed when these fields start up.

Goliat, Skarv and Knarr are examples of projects which spent significantly longer than planned in the execution phase, and where regularity was been low in the first two years after coming on stream. Various types of work and system testing have been needed on these fields after starting production.

4.5 Changes to reserves

Being able to produce oil and gas resources in accordance with or better than the plans is a very important indicator of success for a development project. In the PDO, the operator describes expectations for resources in place and recoverable reserves. These estimates are given as an expected value with an uncertainty range from low to high.

A study from the University of Stavanger has compared actual production during the first four years on stream with the operator’s PDO estimates /8/. This work covers 56 oil developments on the NCS from 1995-2017. One conclusion is that production from the projects during their early years on stream is for the most part overestimated, and that only 25 per cent of developments have ended up with output inside their uncertainty range during the first four years.

At the same time, the NPD’s data reveal that many fields will yield more over their producing lives than the amount which formed the basis for their development. The Resource report for 2019 /4/ shows that field reserves have increased substantially in recent years.

Generally speaking, there has been a trend for reserves to rise in the larger fields and decline for smaller ones. The resource report notes that there could be several explanations for this. Licensees with large discoveries often take a development decision based on the resources needed for profitability, and a flexibility is built in which allows additional resources to be realised over time.

About a third of the 66 projects covered in this report result from the licensees seeing opportunities to initiate measures which yield additional resources in producing fields. An example is Vigdis North-East, the third PDO on the Vigdis subsea development.
About two-thirds of the projects involve developing new fields – in other words, the first PDO for the discovery. Figure 14 presents reserve changes since the PDO on these fields, which are divided into large, medium-sized and small. Several of the small fields have seen a substantial percentage decrease in their post-PDO reserves. One example is Maria, where production experience and data acquisition have established that volumes in place and reservoir properties differ from those described in the PDO. The licensees are working on measures which could increase reserves from the reduced 2018 estimate.

Figure 14 Reserve changes (million standard cubic metres of oil equivalent – scm oe) compared with the PDO for projects covered by the report. Only those representing the first PDO on the field are included (not further development and improved recovery projects). Johan Sverdrup is excluded because of its size (up by 130 million scm oe when phase 2 was sanctioned). Reserve changes based on historical production and remaining reserves on the 2018 resource accounts.

Drilling many appraisal wells before a PDO is often not considered beneficial on small fields. This means that the decision base may be relatively more uncertain than for larger discoveries. It is therefore important that licensees planning to develop small discoveries maximise data acquisition from production wells and other sources in order to improve understanding and reduce uncertainty. A possible measure could be to reduce the number of predrilled wells, and instead drill some of the wells after production has begun.